VoxEU Column Gender Labour Markets

Immigration, elderly care and labour-force participation: Can immigration help women retire later?

Elderly people assisted by immigrant carers, rather than by their sons and daughters, has become a common feature of many European countries. This column presents evidence from Italy suggesting that the immigrant presence in the home-care sector has allowed women, especially those with elderly parents, to retire from their jobs later. Increasing the retirement age has to happen over the coming decades to ensure the sustainability of developed countries’ pension systems.

During the last decade immigrants have increased their presence in the labour force of many rich countries. In several of those countries manually intensive occupations, such as those in the household service sector, have employed many of them. Particularly in Italy, immigrants have disproportionately staffed the long-term care sector for elderly people. The demand from that sector has grown substantially due to the needs of an ageing population. In a recent study (Peri, Romiti and Rossi, 2013) we analyse how immigrants have replaced women in caring for older family members. In particular we measure how this has contributed to increasing women’s planned retirement age. Our empirical strategy considers retirement plans of Italian workers as revealed by national surveys and how they respond to the presence of immigrants in local labour markets. We focus on the difference between men and women and, among those groups, between workers with and without elderly parents. In this way, we are able to detect whether having an elderly parent to care for generates different plans of retirement and if those plans are affected by the presence of local immigrants, who could supply caretaker services. In a country such as Italy where the female labour-force participation and retirement age are low immigrants can represent a valuable market-based mechanism to increasing both of them.

The impact of immigration on the labour supply of women has been investigated before (see, for instance, Barone and Mocetti 2011, Cortès and Tessada 2011, Cortès and Pan 2013, Farrè et al. 2011). The consensus of previous studies is that immigrants, particularly the low-skilled ones, increased the availability and decrease the price of labour in the home-care sector and, as a consequence, affect the labour-supply decisions of women, by increasing their willingness (and decreasing their cost) of working outside of the home. Those studies, however, considered young or working age women, therefore excluding older women who face the biggest burden of elderly parents’ care. In OECD countries the ageing process has brought about a rising demand for long-term care, and in traditional countries such as Italy women provided most of those services in the family (Burda et al. 2008). Ultimately the need to care for elderly parents reduced the already low labour-force participation of women between the age of 45 and 60 (see, e.g., Bolin et al. 2008, Carmichael and Charles 2003, Crespo and Mira 2010, Kolodinsky and Shirey 2000, Marenzi and Pagani 2008). The contribution of immigrants to household service can be particularly valuable for those women with needs to care for their elderly parents.

Immigrants as mechanism to raise retirement age

In our working paper we analyse whether the growing presence of immigrants has affected women’s choices of retirement and labour supply. More specifically, we study whether the planned retirement age of women has increased in those regions with a large increase in immigrant employment relative to other regions. We also analyse whether the planned retirement age increase of women differs relatively to that of men, and we distinguish between women with and without old parents.

Italy represents an ideal case to study this question. Immigration has grown substantially over the last decade, as seen in Table 1, Column 1. Immigrants as a share of the population have increased by almost five percentage points over an eight-year period. At the same time the household service sector has become totally dominated by immigrants. In 2008 they represented almost 80% of the workers in this sector, up from a value of 50% in 2000 (see Table 1, Column 2).

Table 1.

Notes: authors’ calculations from Registry data (first column) and Social Security Archive (INPS) data (second column). Resident immigrants as share of the regional population (first column), immigrant workers as a share of total workers in the domestic sector registered at Social Security Institute (INPS).

During the same years women’s labour-force participation and retirement age increased as well, especially for those with old parents. Table 2 reports summary statistics from Survey on Household, Income and Wealth data. It shows that older women (55 and older) experienced an increase in their planned retirement age and in their probability of working in comparison to similar men during the period 2000-08. This is particularly true for women with old parents.

Table 2.

Notes: authors calculation from Survey on Household, Income and Wealth survey. Planned retirement age and labour supply by gender and living old parents. See Peri et al (2013) for details on the sample.

Our econometric analysis, using the Survey of Household Income and Wealth, shows that raising the presence of immigrants as a share of the regional population increased substantially the planned retirement age and the probability of working for women with respect to men in similar conditions. In particular this increase was higher for women with old parents. Our results are robust to different empirical specifications that control for the other determinants of women’s planned retirement age and for the non-random allocation of immigrants into local areas. We also correct our measure of immigration to account for undocumented immigrants (as revealed by the 2002 amnesty). We found that the results are robust to such checks.

Due to immigration, women older than 45 postponed their planned retirement age by two months compared to men in similar conditions, and if older than 55 and with old parents the delay was twice as long. The increase in immigration (4.7 percentage points of the population) observed between 2000 and 2008 was responsible for a four month increase in planned retirement age for older women, thus explaining 20% of the 1.7 year increase show in the data of Table 2.

Also the probability of working for older women increased relative to similar men, as a consequence of immigration. This effect was more pronounced for older women with elderly parents. The rise of immigrants over the period 2000-2008 has been responsible for an increase in labour supply by nine percentage points for women 55 and older, explaining up to 39% of the increase shown in Table 2.

Women in low-income families show the biggest effects

We also find that the positive effect of immigration on planned retirement age is significant for both women in more and less wealthy families. Splitting the sample, however, we find that those living in households with lower than average-income/education experience the biggest response of planned retirement age to immigration. Women in richer households are probably less affected by the availability of inexpensive elderly care, and they respond less to the cost of care in their retirement and labour-supply decisions. Instead, less wealthy women are more sensitive to the prices of these household services. Hence immigration making elderly care affordable allows them to stay longer in the labour market.


Immigration can be, therefore, a valuable market-based mechanism to rise planned retirement age of older women. In a phase of mandatory increases in the retirement age, as experienced in many OECD countries, to ensure sustainability of the pension system, immigrants can provide a welcome support. In countries such as Italy with strong family ties and a prevalence of women assisting elderly parents, an increase in the local presence of immigrants can help them to stay longer in the labour force. The increase in the retirement age can be, therefore, less painful and pursued through individual incentives if more working immigrants are allowed in the country.


Barone, G and Mocetti, S (2011), “With a little help from abroad: The effect of low-skilled immigration on the female labour supply”, Labour Economics, 18(5), 664–675.

Bolin, K, Lindgren, B, and Lundborg, P (2008), “Your next of kin or your own career?: Caring and working among the 50+ of Europe”, Journal of Health Economics, 27(3), 718–738.

Burda, M, Hamermesh, D, and Weil, P (2008), “The distribution of total work in the EU and in US”, in Boeri, T, Burda, M and, Kramarz F (eds.) Working Hours and Job Sharing in the EU and USA: Are Europeans Lazy? Or Americans Crazy?, Oxford University Press.

Carmichael, F and Charles, S (2003), “The opportunity costs of informal care: does gender matter?”, Journal of Health Economics, 22(5), 781–803.

Cortès, P and Pan, J (2013), “Outsourcing household production: Foreign domestic workers and native labour supply in Hong Kong”, Journal of Labor Economics, 31(2), 327–371.

Cortès, P and Tessada, J (2011), “Low-skilled immigration and the labour supply of highly skilled women”, American Economic Journal: Applied Economics, 3(3), 88–123.

Crespo, L and Mira, P (2010), “Caring for parents and employment of European mature women”, Working Papers wp2010_1007, CEMFI.

Farrè, L, Gonzalez, L, and Ortega, F (2011), “Immigration, family responsibilities and the labour supply of skilled native women”, B.E. Journal of Economic Analysis and Policy: Contributions to Economic Analysis and Policy 11(1), 1–48.

Kolodinsky, J and Shirey, L (2000), “The impact of living with an elder parent on adult daughter’s labour supply and hours of work”, Journal of Family and Economic Issues, 21, 149–175.

Marenzi, A and L Pagani (2008), “The labour market participation of sandwich generation Italian women”, Journal of Family and Economic Issues, 29(3), 427-444.

Peri, G, Romiti, A, and Rossi, M (2013), “Immigrants, Household Production and Women’s Retirement”, IZA Discussion paper n. 7549.

2,100 Reads